VOLUME XLV 



NUMBER 6 



Botanical Gazette 



JUNE igoS 



RELATION OF MEGASPORES TO EMBRYO SACS IN 



ANGIOSPERMS 



CONTRIBUTIONS FROM THE HULL BOTANICAL LABORATORY 110 



John M. Coulter 



The recorded cases of unusual conditions in the angiosperm embryo 

 sac are multiplying. Especially is this true in reference to the occur- 

 rence of more than eight nuclei in the ante-fertilization development 

 of the gametophyte. There is a tendency to regard the latter cases 

 as representing a more primitive condition of the gametophyte, and to 

 exclude from consideration a general view of the genesis of the nuclei. 

 It seems to be a fitting time to call attention to certain facts that may 

 interpret this situation, may connect it with other "irregularities," 

 may check the tendency to ill-considered generalization, and may 

 indicate certain things that must be included in any investigation of 

 the embryo sac. 



In the genesis of the ordinary angiosperm embryo sac from the 

 megaspore mother cell five successive nuclear divisions are involved, 

 the first two being the reduction divisions (usually resulting in four 

 megaspores with walls), and the other three resulting in the produc- 

 tion of eight nuclei from the nucleus of one of the megasporcs. The 

 functions of these eight nuclei of the female gametophyte do not enter 

 into the present discussion. Attention is called to the following facts: 

 this genesis begins with the division of the mother cell; the essential 

 part of the process is found in the first two divisions, which cannot be 

 omitted, so far as we know, when fertilization is to occur ; the result 

 of these first two divisions is the organization of megaspore nuclei, 

 definitely recognized by their cytological history and structure. 



The most famiHar illustration of departure from this ordinary 

 sequence is the case of Lilium and its allies. In current description 



361 



